Voice enhancement method based on DNN noise classification

Abstract

The invention provides a voice enhancement method based on DNN noise classification, and relates to the technical field of voice recognition. The method comprises the following steps: carrying out preprocessing on noise signals, and determining signals of the non-voice segment; for each frame of voice signals, carrying out fast Fourier transform, and calculating the spectral line energy; enablingthe frequency spectrum of the voice signals to pass through a Mel filter group, multiplying the frequency spectrum energy with the frequency response of the Mel filters, thus obtaining the Mel filtering energy; carrying out discrete cosine transform on the logarithm of the Mel filtering energy of each frame of signals, thus obtaining a Mel cepstrum parameter which serves as the feature vector of each frame of Mel filtering of the voice; taking the feature vector of each frame of Mel filtering as a 24-dimensional vector, and also as the input of the deep neural network; and carrying out training and classifying on the noises by utilizing the deep neural network. For the voice enhancement method based on DNN noise classification, through the classification for noises, the subsequent voice enhancement quality is obviously improved in the subjective / object testing.

Description

technical field [0001] The invention relates to the technical field of speech recognition, in particular to a speech enhancement method based on DNN noise classification. Background technique [0002] Speech enhancement is a technology that extracts useful speech signals from background noise, and suppresses and reduces noise interference after speech signals are interfered by various noises. Speech enhancement is an effective method to solve the noise pollution of speech signals. It is also a key link and step in speech signal processing, and is widely used in people's production and life. The interference of background noise deteriorates the performance of speech signal processing, such as speech coding, speech synthesis, and speech recognition. For example, speech recognition is a key step in human-computer interaction using speech signals. The existing speech recognition system has a high recognition rate in a quiet environment, but in a strong noise environment, the r...

AI Technical Summary

Problems solved by technology

The interference of background noise deteriorates the performance of speech signal processing, such as speech coding, speech synthesis, speech recognition, etc.

For example, speech recognition is a key step in human-computer interaction using speech signals. The existing speech recognition system has a high recognition rate in a quiet environment, but in a strong noise environment, the recognition rate will drop rapidly, which cannot meet the requirements of practical applications. need

Images